Skip to main content

​
Overview

Historia Diaria is an automated story generator that runs daily using GitHub Actions. It combines AI-powered content generation with static HTML output to create a self-updating story archive.

​
System Architecture

The system consists of four main components:

Python Script

Core logic for API calls, content generation, and file management

GitHub Actions

Automated scheduler that runs the script daily at 8:00 AM Lima time

HTML Template

Reusable layout with dynamic content placeholders

JSON History

Persistent storage for story metadata and navigation

​
Workflow Diagram

The automated workflow follows this sequence:

​
Automated Scheduling

The GitHub Actions workflow is configured to run automatically:
actualizar.yml
name: Actualizar Historia Diaria

on:
  schedule:
    # UTC time: 13:00 = 8:00 AM Lima time
    - cron: '0 13 * * *'
  workflow_dispatch: # Manual trigger option

jobs:
  generar-pagina:
    runs-on: ubuntu-latest
    permissions:
      contents: write
The workflow_dispatch trigger allows you to manually run the workflow from the GitHub Actions tab for testing.

​
API Integration Flow

The script connects to OpenRouter’s API to generate stories:
generar_historia.py
# 1. API Connection
api_key = os.environ.get("OPENROUTER_API_KEY")

response = requests.post(
    url="https://openrouter.ai/api/v1/chat/completions",
    headers={
        "Authorization": f"Bearer {api_key}",
        "Content-Type": "application/json"
    },
    data=json.dumps({
        "model": "stepfun/step-3.5-flash:free",
        "messages": [{"role": "user", "content": prompt}]
    })
)

respuesta_ia = response.json()['choices'][0]['message']['content']
The API key is stored as a GitHub Secret (OPENROUTER_API_KEY) and passed as an environment variable during workflow execution.

​
File Generation Process

The script generates two HTML files on each run:
1

Generate Daily Filename

Creates a date-stamped filename using the format historia-YYYY-MM-DD.html
generar_historia.py
fecha_hoy = datetime.now()
año = fecha_hoy.strftime("%Y")
mes_num = fecha_hoy.strftime("%m")
dia = fecha_hoy.strftime("%d")

nombre_archivo_hoy = f"historia-{año}-{mes_num}-{dia}.html"
2

Fill Template with Content

Replaces placeholder variables with generated content
generar_historia.py
with open('plantilla.html', 'r', encoding='utf-8') as archivo:
    contenido_html = archivo.read()

contenido_html = contenido_html.replace('{{TITULO}}', nuevo_titulo)
contenido_html = contenido_html.replace('{{HISTORIA}}', nueva_historia)
contenido_html = contenido_html.replace('{{IMAGEN_URL}}', url_imagen)
contenido_html = contenido_html.replace('{{MENU}}', menu_html)
3

Save Both Files

Writes index.html (latest story) and the dated archive file
generar_historia.py
# Latest story (homepage)
with open('index.html', 'w', encoding='utf-8') as archivo:
    archivo.write(contenido_html)

# Archived version
with open(nombre_archivo_hoy, 'w', encoding='utf-8') as archivo:
    archivo.write(contenido_html)

​
History Management

The historial.json file maintains a structured archive of all generated stories:

​
Data Structure

historial.json
{
    "Marzo 2026": [
        {
            "titulo": "El Guardián del Faro de las Sombras",
            "archivo": "historia-2026-03-04.html"
        },
        {
            "titulo": "El Código de las Estrellas",
            "archivo": "historia-2026-03-02.html"
        }
    ],
    "Febrero 2026": [
        {
            "titulo": "El Eco de los Días Perdidos",
            "archivo": "historia-2026-02-26.html"
        }
    ]
}

​
Update Logic

The script implements smart update logic to handle re-runs on the same day:
generar_historia.py
# Load existing history
if os.path.exists('historial.json'):
    with open('historial.json', 'r', encoding='utf-8') as f:
        historial = json.load(f)
else:
    historial = {}

# Create month key if new
if llave_mes not in historial:
    historial[llave_mes] = []

# Update if exists, insert if new
historia_actualizada = False
for item in historial[llave_mes]:
    if item['archivo'] == nombre_archivo_hoy:
        item['titulo'] = nuevo_titulo  # Update existing
        historia_actualizada = True
        break

if not historia_actualizada:
    # Insert at beginning (newest first)
    historial[llave_mes].insert(0, {
        "titulo": nuevo_titulo,
        "archivo": nombre_archivo_hoy
    })
This logic prevents duplicate entries when the workflow runs multiple times on the same day, updating the title instead of creating a new entry.

​
Error Handling

The script includes fallback mechanisms for API failures:
generar_historia.py
try:
    response = requests.post(
        url="https://openrouter.ai/api/v1/chat/completions",
        headers={"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"},
        data=json.dumps({"model": "stepfun/step-3.5-flash:free", "messages": [{"role": "user", "content": prompt}]})
    )
    respuesta_ia = response.json()['choices'][0]['message']['content']
except Exception as e:
    respuesta_ia = "<TITULO>Error</TITULO><HISTORIA>Fallo al conectar con la API.</HISTORIA><IMAGEN>error</IMAGEN>"
If the API call fails, a default error message is used to ensure the workflow completes successfully.

​
Next Steps

Story Generation

Learn how AI generates and parses story content

HTML Templates

Explore the template structure and styling system

Build docs developers (and LLMs) love

Get started for freeTalk to us